Improving bright-field images for tracking particles with Defocustracker

Sammanfattning: Acoustophoresis is a cell separation technology where particles or cells are suspended in a fluid container and subject to acoustic standing waves. The standing waves exerts an acoustic radiation force on the particles, pushing them towards the nodal plane. This movement is called migration, and the migration velocity depends on the acoustic mobility of the particle, which depend on the particle size, density and compressibility relative to the suspension fluid. Acoustophoresis is used in biomedical and clinical applications, such as diagnostics.\\ The velocities of the particles migrating in the container are determined by individually tracking the particles in a software called Defocustracker. A series of images is taken of the migration and are then analyzed in Defocustracker. In order to perform the particle tracking, the images must be dark-field and the particles fluorescent - in bright-field mode Defocustracker cannot properly differentiate between particles and noise in the images, as bright-field introduces a lot of background noise. There is a concern that dying the particles may affect their properties, which is why it would be preferable to conduct the acoustophoresis in bright-field. In some cases, the cells are not even possible to dye, which is another motivation to track cells in bright-field. The aim of this candidate thesis is to find a pre-processing algorithm that is able to improve the images, so that the particle tracking can be done in bright-field.